Attenuation of smoke-induced lung by treatment with a soluble hydrolase inhibitor

Kevin R. Smith*, Kent E. Pinkerton*, Takaho Watanabe†, Theresa L. Pedersen†, Seung Jin Ma†, and Bruce D. Hammock†‡

*Center for Health and the Environment and †Department of Entomology and Cancer Research Center, University of California, Davis, CA 95616

Contributed by Bruce D. Hammock, December 21, 2004 Changes in the lungs due to smoking include inflammation, epi- small and medium pulmonary vessels (11). Epoxyeicosatrienoic thelial damage, and remodeling of the airways. Airway inflamma- acids (EETs) are metabolites of arachidonic acids that undergo tion is likely to play a critical role in the genesis and progression of hydrolysis by sEH to form dihydroxyeicosatrienoic acids tobacco smoke-induced airway disease. Soluble (DHETs). There is evidence that EETs can prevent vascular (sEH) is involved in the metabolism of endogenous chemical inflammation through inhibition of expression of several cell mediators that play an important role in inflammation. Epoxyei- adhesion molecules (CAMs) including E-selectin, vascular cell cosatrienoic acids (EETs) have demonstrated antiinflammatory adhesion molecule 1, and intercellular adhesion molecule 1 (12) properties, and hydrolysis of these by sEH is known to and by preventing leukocyte adhesion to the vascular wall diminish this activity. To examine whether acute tobacco smoke- (12–14). The role of sEH and EETs in tobacco smoke-induced induced inflammation could be reduced by a sEH inhibitor, 12-(3- inflammation and lung disease has not been investigated. adamantane-1-yl-ureido)-dodecanoic acid n-butyl ester was given sEH has been identified as a promising pharmacological by daily s.c. injection to spontaneously hypertensive rats exposed target, and inhibition of sEH has been proposed as a novel ͞ to filtered air or tobacco smoke for a period of 3 days (6 h day). approach for the treatment of diseases, including Acute exposure to tobacco smoke significantly increased by 3.2- (15, 16). We have previously established a model of acute fold (P < 0.05) the number of cells recovered by bronchoalveolar inflammation in the spontaneously hypertensive (SH) rat (17). lavage. The sEH inhibitor significantly decreased total bronchoal- We tested in this model whether inflammation associated with veolar lavage cell number by 37% in tobacco smoke-exposed rats short-term exposure to tobacco smoke could be altered with the with significant reductions noted in neutrophils, alveolar macro- use of a sEH inhibitor, 12-(3-adamantane-1-yl-ureido)- phages, and lymphocytes. A combination of sEH inhibitor and EETs dodecanoic acid n-butyl ester (AUDA-nBE), in the absence or was more significant in its ability to further reduce tobacco smoke- presence of EETs. induced inflammation compared with the sEH inhibitor alone. The sEH inhibitor led to a shift in some plasma epoxides and diols that Materials and Methods are consistent with the hypothetical action of these compounds. Reagents, Chemicals, and Analytical Procedures. We conclude that an sEH inhibitor, in the presence or absence of AUDA-nBE and EETs, can attenuate, in part, inflammation associated with acute 1-cyclohexyl-3-tetradecyl urea, 1-phenyl-2-hexanoic acid urea, exposure to tobacco smoke. 20-hydroxyeicosanoic acid, and 1-cyclohexyl-3-dodecanoic acid urea were synthesized in our laboratory. These products were epoxyeicosatrienoic acids ͉ pulmonary ͉ antiinflammatory purified by recrystallization and characterized structurally by [1H]NMR and͞or [13C]NMR, infrared, and mass spectroscopy. LC-MS͞MS analysis was performed by using a Micromass igarette smoking is associated with a number of pulmonary Quattro Ultima triple quadrupole tandem mass spectrometer diseases including bronchitis, airway obstruction, and em- C (Micromass, Manchester, U.K.) described in Supporting Text, physema. Collectively these pulmonary maladies are referred to which is published as supporting information on the PNAS web as chronic obstructive pulmonary disease (COPD). COPD is site. Conditions for pharmacokinetic analysis are described in prevalent in Ϸ20 million men and women in the United States Supporting Text. As presented in detail in Supporting Text, EETs and is the fourth leading cause of death (1). The pathology of chronic bronchitis and COPD includes airway mucus gland were synthesized by using acrachidonic acid methylester and hyperplasia, mucous hypersecretion, and an influx of inflamma- m-chloro-perbenzoic acid, and the regioisomer ratio was deter- tory cells including neutrophils, , and lymphocytes mined with LC-MS based on authentic pure standards purchased (2–6). The genesis of this disease is thought to lie in the from Cayman Chemical (Ann Arbor, MI). The composition of inflammatory process induced by tobacco smoke, leading to cell the mixture was 10% of 8(9)-EET, 40% of 11(12)-EET, and 50% injury, cellular hyperplasia, and occasionally neoplasia. There- of 14(15)-EET. fore, it is important for us to understand the process by which tobacco smoke induces inflammation in the lungs. Oxylipid Analysis. EETs, DHETs, DiHOMEs, and linoleate ep- ͞ Soluble epoxide hydrolases (sEH) are enzymes that add water oxides (EpHOMEs) were analyzed by LC-MS MS by using to epoxides forming their corresponding 1,2-diols (7, 8). Diols of negative mode electrospray ionization with a tandem mass linoleate epoxides (DiHOMEs) have been implicated in inflam- spectral detector (Quattro Ultima, Micromass) operated in a matory disorders, including adult respiratory distress syndrome multireaction monitoring mode as described in refs. 18 and 19. (9), and may be endogenous regulators of vascular permeability and inflammation (10). Histopathologic evaluation of Swiss– Webster mouse lung after dosing with 12,13-DiHOME show Abbreviations: AUDA, 12-(3-adamantan-1-yl-ureido)-dodecanoic acid; AUDA-nBE, AUDA n-butyl ester; BAL, bronchoalveolar lavage; BALF, BAL fluid; COPD, chronic obstructive massive alveolar edema and hemorrhage with interstitial edema pulmonary disease; DiHOME, diols of linoleate epoxide; DHET, dihydroxyeicosatrienoic in vessel walls of the lung. A 50% mortality was observed at a acid; EET, epoxyeicosatrienoic acid; EpHOME, linoleate epoxide; sEH, soluble epoxide dose of 100 mg͞kg within 6.5 h of administration (through the hydrolase; SH, spontaneously hypertensive. tail vein). Immunohistochemistry of the lung tissue showed ‡To whom correspondence should be addressed. E-mail: [email protected]. epoxide hydrolase concentrated locally in the smooth muscle of © 2005 by The National Academy of Sciences of the USA

2186–2191 ͉ PNAS ͉ February 8, 2005 ͉ vol. 102 ͉ no. 6 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0409591102 Downloaded by guest on October 2, 2021 The acquired data were quantified with the MASSLYNX software s.c. Implantation of EETs for Tobacco Smoke Exposures. Wax formu- package (Micromass). lations containing EET regioisomers were implanted s.c. 1 day before onset of exposure to filtered air or tobacco smoke. The Formulation of EETs Wax Plug. To create a wax plug,5gofbeeswax total dose of EETs in the wax plug implant was 2.5 mg͞kg of body (Aldrich) was melted at 100°C for 20 min by using a hot plate, weight. Four animals from the control group and four animals and EETs (0.5 mg) were added to the molten wax while stirring. from the tobacco smoke-exposed group were implanted with the The wax-EETs suspension was then poured into a mold made slow-release formulation of EETs. The approach of a single s.c. with glass plates and cooled to room temperature. The resultant implantation for the 3-day study was selected to minimize stress wax stick containing EETs was cut to suitable size. No degra- to animals from anesthesia. dation of EETs was observed during this process. Release rates of EETs from the wax plug in vitro were investigated as modified s.c. Injection of AUDA-nBE for Tobacco Smoke Exposures. AUDA- from a reported method (20), with details of the method nBE (7 mg͞ml corn oil) was s.c. administered in SH rats at a dose presented in detail in Supporting Text. of 10 mg͞kg of body weight. The total volume injected was between 0.36 and 0.46 ml. Animals were injected with AUDA- Animals. Healthy, 11-week-old male SH (SHR͞NCrlBR) rats nBE each day before exposure to tobacco smoke. Doses of (derived from WKY rats by phenotypic segregation of the AUDA-nBE used in this study were selected based on results hypertensive trait and inbreeding) were purchased from Charles from preliminary pharmacokinetic studies in mice and rats. River Laboratories (Portage, MI) and quarantined for 1 week These doses were selected to provide optimal efficacy and before exposure to tobacco smoke. Animals were handled in minimal toxicity over a 3-day period. Four animals from the accordance with standards established by the U.S. Animal control group and four animals from the tobacco smoke-exposed Welfare Acts set forth in National Institutes of Health guidelines group were injected with AUDA-nBE. Four animals with EET and the University of California, Davis, Animal Care and Use implants from the control group and four animals with EET Committee. The rats were housed in plastic cages with TEK- implants from the tobacco smoke-exposed group were injected Chip pelleted paper bedding (Harlan Teklad, Madison, WI) and with AUDA-nBE. In addition, four control animals and four maintained on a 12 h light͞12 h dark cycle. All animals had tobacco smoke-exposed animals were injected with corn oil by access to water and Laboratory Rodent Diet 5001 purchased using the same protocol as animals injected with AUDA-nBE. from LabDiet (Brentwood, MO) ad libitum before, during, and after exposures. Tobacco Smoke Exposure. Rats were exposed to a mixture of sidestream and mainstream cigarette smoke in a smoking appa- Treatment of Animals for Pharmacokinetics Study. Animals were ratus built in our laboratory (21). The cigarettes were humidified selected for pharmacokinetic studies based on a body weight 2R4F research cigarettes (Tobacco Health Research Institute; stratified randomization procedure after a 1- to 2-week accli- Lexington, KY). An automatic metered puffer was used to mation period. The body weight of animals was 250–280 g. A 10 smoke the cigarettes under Federal Trade Commission condi- mg͞kg of body weight dosing of these inhibitors (7 mg͞ml corn tions (35-ml puff, 2-sec duration, 1 puff per min). The smoke was oil) were s.c. administered to SH rats. The animals were injected collected in a chimney, diluted with filtered air, and delivered to

one time with AUDA-nBE, and blood concentrations of sEH whole-body exposure chambers. The exposures were character- PHARMACOLOGY inhibitor(s) were determined in the experimental animals after ized for three major constituents of cigarette smoke: , 72 h of exposure to filtered air. Plasma (collection described , and total suspended particulates (TSP). An- below) concentrations of sEH inhibitor(s) were determined in imals were exposed for 6 h͞day for 3 days. Carbon monoxide was animals separate from those used in the above study after measured every 30 min, total suspended particulates every 2 h, exposure to tobacco smoke or filtered air for 3 days with the and nicotine daily (about midway through the exposure period). animals treated daily with s.c. injections of AUDA-nBE (three Data for smoke exposure characteristics are presented as total injections). Blood from these animals was drawn only at mean Ϯ SD. necropsy (immediately after exposure on the third day of the study). Bronchoalveolar Lavage (BAL). Established protocols were fol- lowed for BAL of animals (22). Eighteen hours after the last Blood Sample Preparation. After administration, serial tail-bled exposure to tobacco smoke, animals were anesthetized with an blood samples (Ͻ10 ␮l) were collected at various time points (30 overdose of sodium pentobarbital. This timing was used to min to 72 h). Blood samples were transferred to 1.5-ml Eppen- ensure a robust inflammatory response. The trachea was can- dorf microcentrifuge tubes. Blood samples were weighed with an nulated and the lung lavaged with one aliquot of Ca2ϩ͞Mg2ϩ- analytical balance and vortexed with 100 ␮l of purified water and free PBS (pH 7.4). The volume of the aliquot was equal to 35 25 ␮l of internal standard in ethyl acetate (500 ng͞ml 1-cyclo- ml͞kg of body weight (Ϸ90% of total lung capacity). The aliquot hexyl-3-tetradecyl urea). The samples were extracted with 500 ␮l was instilled into the lungs three times before final collection. of ethyl acetate. The ethyl acetate layer was transferred to a The BAL fluid (BALF) was immediately centrifuged at 250 ϫ 1.5-ml Eppendorf microcentrifuge tube and dried under nitro- g for 10 min at 4°C to remove cells. The cell pellet was then gen. The residue was reconstituted in 25 ␮l of methanol. resuspended in PBS, and the cells were counted with a hemo- Aliquots (10 ␮l) were injected onto the LC-MS͞MS system. cytometer. Cell differentials were performed on cytospin prep- arations (Shandon, Pittsburgh) stained with Hema 3 (Fisher Pharmacokinetics Analysis. Pharmacokinetic parameters were ob- Scientific). Macrophages, neutrophils, lymphocytes, and eosin- tained by fitting the blood concentration–time data to a non- ophils were counted by using light microscopy (1,000 cells per compartmental model with WINNONLIN software (Pharsight, sample). Mountain View, CA). Parameters estimated included lambda z (␭z), time of maximum concentration (Tmax), maximum con- Plasma. Additional rats were exposed to filtered air or tobacco centration (Cmax), elimination half-life (T1/2), area under the smoke for 3 days with treatments and exposures as described concentration–time curve to terminal time (AUCt), area under above, and blood was collected at necropsy for measurement of the concentration–time curve to infinite time (AUCϱ) and the levels of oxylipids and sEH inhibitor(s) in plasma. Immediately mean residence time (MRT). AUCt was calculated by the linear͞ after the last exposure to tobacco smoke, animals were anes- log trapezoidal rule. thetized with an overdose of sodium pentobarbital, and blood

Smith et al. PNAS ͉ February 8, 2005 ͉ vol. 102 ͉ no. 6 ͉ 2187 Downloaded by guest on October 2, 2021 was collected from the caudal vena cava and placed in Vacu- tainer tubes containing EDTA. This timing was used to ensure that epoxy and diol levels were above limits of detection for the instrumentation. The tubes were placed on ice until collection was completed, then centrifuged at 2,000 ϫ g for 20 min for plasma separation. Plasma was prepared as follows for oxylipid analysis. Eppendorf tubes (2 ml) were spiked with 10 ␮lof antioxidant solution [0.01 mg of butylated hydroxytoluene (BHT) and EDTA], and 1.5 ml aliquots of plasma were por- tioned into the tubes and immediately stored at Ϫ80°C until oxylipid analysis.

Extraction of Oxylipids from Plasma. Plasma samples were thawed to room temperature. Aliquots (250 ␮L) were spiked with 10 ␮l of surrogate solution containing dihydroxynonadecanoic acid and epoxyheptadecanoic acid, then extracted by 60 mg of Oasis HLB (Waters) solid phase extraction cartridges (SPE) according to the following regimen: SPE were washed and preconditioned with 2 ml of HPLC-grade methanol (VWR Scientific) and 2 ml of 2.5 mM H2PO4 with 10% methanol (pH 3.8). Samples were loaded and 250 ␮l of the phosphoric acid͞10% methanol was Fig. 1. Blood concentration–time profiles of AUDA-nBE and AUDA in SH rats after s.c. administration of 10 mg͞kg AUDA-nBE. Data represent the mean Ϯ added. Samples were allowed to extract by gravity, followed by ϭ a 2-ml wash with the phosphoric acid͞methanol solution. Car- SD (n 3 animals). Inset shows structures of the parent (AUDA-nBE) and major active metabolite (AUDA). PK parameters were obtained by fitting the blood tridges were air-dried 15 min by gentle vacuum and then eluted concentration–time data to a noncompartmental model. PK parameters are as by gravity drip with 2 ml of ethyl acetate (VWR Scientific). follows: ␭z (1͞h), 0.0309; Tmax (h), 2.00; Cmax (nM), 136.6; T1/2 (h), 22.4; area Eluate was brought to dryness by evaporation under a gentle under the concentration–time curve to terminal time (AUCt) (nM⅐h), 4,540; blanket of nitrogen and reconstituted in 100 ␮l of methanol mean residence time (hr), 25.1. The concentration of AUDA after 72 h was 37.3 containing 1-phenyl-2-hexanoic acid urea, 20-hydroxyeicosanoic nM after a single s.c. administration. AUDA concentrations after 3 days of acid, and 1-cyclohexyl-3-dodecanoic acid urea as internal stan- exposure to filtered air or tobacco smoke in rats treated with daily (total of dards for oxylipid analysis by LC-MS͞MS. three) s.c. injections of AUDA-nBE were as follows (mean Ϯ SD): filtered air, AUDA-nBE, and EETs, 208 Ϯ 16 nM; filtered air and AUDA-nBE, 152 Ϯ 23 nM; tobacco smoke, AUDA-nBE, and EETs, 298 Ϯ 126 nM; tobacco smoke and Data Analysis. All numerical BAL data were calculated as mean Ϯ AUDA-nBE, 325 Ϯ 117 nM (the last two values had a single high value). SD. Comparisons between groups were made by analysis of variance followed by Fisher’s protected least significant differ- Ͻ ence posttest. A P 0.05 was considered significant. Statistical AUDA-nBE and EETs before tobacco smoke exposure. The analysis was performed with STATVIEW 5.0.1 (SAS Institute, Cary, number of BALF macrophages was increased significantly after NC). Comparisons of mean oxylipid concentrations in plasma 3 days of tobacco smoke exposure (Fig. 2B) but this result was were made between groups by using a two-tailed t test, assuming significantly reduced by AUDA-nBE given before smoke expo- Ͻ equal variance, with P 0.05 considered significant. sure. Treatment with both AUDA-nBE and EETs did not Results further decrease number (Fig. 2B). Neutrophil numbers in BALF were also significantly increased after 3 days Release Rate of EETs from Wax Plug in Vitro. The release rate of of tobacco smoke exposure (Fig. 2C) but this result was signif- EETs into water from wax plugs was estimated to be 10 Ϯ 0.4 icantly reduced by AUDA-nBE given before tobacco smoke ␮g͞day. exposure. Combined treatment with AUDA-nBE and EETs further reduced the number of neutrophils recovered by lavage. Tobacco Smoke Exposure Characteristics. Total suspended particu- Lymphocytes were significantly increased in BALF after expo- late levels in tobacco smoke during the 3-day study were 76.4 Ϯ sure to tobacco smoke for 3 days (Fig. 2D), but this result was 16.0 mg͞m3, nicotine levels were 6.8 Ϯ 0.2 mg͞m3, and carbon monoxide concentration was 234 Ϯ 2 ppm. significantly decreased by AUDA-nBE treatment before smoke exposure to levels similar to those observed in filtered-air control Pharmacokinetics. To estimate blood concentration of AUDA- animals. Treatment of animals with AUDA-nBE and EETs nBE and 12-(3-adamantane-1-yl-ureido)-dodecanoic acid before exposure to tobacco smoke did not result in further (AUDA) in SH rats, a pharmacokinetic study was performed reduction of BAL lymphocyte numbers compared with treat- with a single dose. Fig. 1 shows blood concentration–time ment with only AUDA-nBE. Numbers of eosinophils were profiles of AUDA-nBE and AUDA in SH rats after s.c. admin- increased in the BALF after 3 days of tobacco smoke exposure istration of 10 mg͞kg AUDA-nBE. AUDA-nBE was rapidly but not to a statistically significant level (Fig. 2E). Of note was metabolized to AUDA, which is an equally potent inhibitor of that the number of total BALF cells and macrophages from rats sEH. Thus, AUDA-nBE is a pro-drug for AUDA to improve treated with AUDA-nBE and EETs before exposure to filtered bioavailability, although both compounds show a low nanomolar air was significantly lower compared with those recovered from Ki with the purified recombinant sEH. animals treated only with vehicle before exposure to filtered air (Fig. 2 A and B). A similar trend was observed for BAL BAL. The volume of BALF recovered was not significantly lymphocytes in animals exposed to filtered air, but statistical different between treatment͞exposure groups. The total number significance was not obtained (Fig. 2D). Cell differentials are of cells in the BALF was increased significantly after 3 days of shown as percentages in Table 1 and exhibit similar trends to the tobacco smoke exposure. s.c. injection of AUDA-nBE before numbers of different cell types recovered in BALF. smoke exposure significantly decreased the number of BALF cells compared with treatment with the vehicle (Fig. 2A), and this Oxylipids in Plasma. 12(13)-EpOME and 9(10)-EpOME (linoleate result was further decreased by treatment of animals with both epoxides), 14(15)-, 11(12)-, 8(9)-, and 5(6)-EET (arachidonate

2188 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0409591102 Smith et al. Downloaded by guest on October 2, 2021 erage recoveries ranged between 94% and 100% with the percent of relative SD between 10% and 14%. The diol and epoxy surrogates had average recoveries of 62 and 98% with the percent of relative SD of 22% and 18%, respectively. Control criteria for precision within 25% was met for surrogate recov- eries. Arachidonate diol and epoxide concentrations in plasma ranged from 1.03 to 5.17 nM and from 0.00 to 5.05 nM, respectively (Table 2). Linoleate diol and epoxide concentra- tions ranged from 11.4 to 84.1 nM and from 3.26 to 35.1 nM, respectively. BALF was also extracted for oxylipids according to plasma protocol, but values for many analytes were near or below detection limits. Further method development is necessary for determining oxylipids in BALF. 12(13)- and 9(10)-EpOME plasma concentrations were sig- nificantly decreased after 3 days of tobacco smoke exposure in animals treated with vehicle (Table 2). 14(15)-, 11(12)-, and 5(6)-EET were decreased in these animals but not significantly. 8(9)-EET was significantly increased with exposure to tobacco smoke. There was also a significant increase in all arachidonate and linoleate diol concentrations in plasma with exposure to tobacco smoke. Administration of AUDA-nBE to rats before tobacco smoke exposure significantly increased 12(13)- and 9(10)-EpOME and 14(15)-EET concentrations in plasma compared with levels in rats exposed to tobacco smoke plus vehicle (Table 2). 11(12)- and 5(6)-EET concentrations were increased but not signifi- cantly in tobacco smoke-exposed rats treated with AUDA-nBE. 12(13)-EpOME, 9(10)-EpOME, and 11(12)-EET were signifi- cantly increased in smoke-exposed animals that received both AUDA-nBE and EETs compared with levels in animals exposed to tobacco smoke and treated only with vehicle. AUDA-nBE alone or in combination with EETs did not significantly change levels of 8(9)-EET or 5(6)-EET in animals exposed to tobacco smoke. Administration of AUDA-nBE to rats before exposure to tobacco smoke significantly decreased 12,13-DiHOME con-

centrations compared with levels in rats treated with vehicle and PHARMACOLOGY exposed to tobacco smoke. In rats that received both AUDA- nBE and EETs before exposure to tobacco smoke, 12,13- and 9,10-DiHOME and 14,15- and 5,6-DHET were significantly decreased compared with levels in tobacco smoke-exposed rats treated with vehicle. 11,12-DHET was significantly increased and 8,9-DHET was not significantly changed in smoke-exposed animals that received both AUDA-nBE and EETs. Discussion There is a considerable amount of research to support a key role for inflammation as a driving force to cause the airway epithe- lium to undergo changes leading to the loss of ciliated cells, hypersecretion of mucin, bronchitis, emphysema, and lung can- cer. Smoking causes a local cytokine secretion in the lung, which leads to an infiltration of leukocytes into the airways and alveolar destruction. We have previously demonstrated the ability of a Fig. 2. Bronchoalveolar lavage cell characteristics after exposure of rats to catalytic antioxidant AEOL 10150 to decrease tobacco smoke- filtered air or tobacco smoke. Numbers of total cells (A), macrophages (B), induced inflammation in the lungs of rats, suggesting a role of neutrophils (C), lymphocytes (D), and eosinophils (E) in BAL from rats exposed to filtered air or tobacco smoke for 3 days. Rats were exposed to filtered air oxygen radicals in the induction of proinflammatory cytokines and chemokines (17), possibly through oxidant-mediated acti- (gray bars) after treatment with vehicle, AUDA-nBE, or AUDA-nBE and EETs. ␬ Additional rats were exposed to tobacco smoke (black bars) after treatment vation of the redox-sensitive transcription factor, NF- B. How- with vehicle, AUDA-nBE, or AUDA-nBE and EETs. Data are presented as ever, inflammation induced by tobacco smoke was not resolved mean Ϯ SD (n ϭ 4). †, P Ͻ 0.05 compared with respective filtered air control. to baseline levels by treatment with the antioxidant, suggesting ‡, P Ͻ 0.05 compared with tobacco smoke and vehicle. §, P Ͻ 0.05 compared a role of additional mediators of inflammation. Corticosteroids with tobacco smoke and AUDA-nBE. ¶, P Ͻ 0.05 compared with filtered air and have antiinflammatory properties, making these compounds vehicle. No eosinophils were observed in BAL from animals treated with useful in the treatment of COPD. A review of several important vehicle before exposure to filtered air or from animals treated with AUDA-nBE studies does not show evidence of significant improvement in the and EETs before exposure to tobacco smoke. symptoms of patients with COPD treated with systemic corti- costeroids (23), suggesting a need for additional treatment epoxides), 12,13-DiHOME and 9,10-DiHOME (linoleate diols), modalities. and 14,15-, 11,12-, 8,9-, and 5,6-DHET (arachidonate diols) sEH is involved in the regulation of EETs and other epoxy- concentrations were analyzed in plasma. Internal standard av- lipids (24, 25) and may therefore be an important mediator of

Smith et al. PNAS ͉ February 8, 2005 ͉ vol. 102 ͉ no. 6 ͉ 2189 Downloaded by guest on October 2, 2021 Table 1. Cell differentials in BAL after 3 days of tobacco smoke exposure in rats Vehicle (corn oil) AUDA-nBE AUDA-nBE ϩ EETs

Filtered air Tobacco smoke Filtered air Tobacco smoke Filtered air Tobacco smoke

Macrophages, % 90.2Ϯ2.9 48.7Ϯ6.9†§ 92.1Ϯ3.2 62.9Ϯ3.6†‡ 93.4Ϯ3.9 72.7Ϯ6.8†‡§ Neutrophils, % 9.0Ϯ2.7 50.7Ϯ6.8†§ 7.3Ϯ3.5 36.8Ϯ3.2†‡ 6.2Ϯ3.7 27.1Ϯ6.8†‡§ Lymphocytes, % 0.80Ϯ0.33 0.60Ϯ0.23§ 0.55Ϯ0.30 0.25Ϯ0.19‡ 0.40Ϯ0.16¶ 0.25Ϯ0.10‡ Eosinophils, % 0.00Ϯ0.00 0.05Ϯ0.10 0.15Ϯ0.19 0.10Ϯ0.20 0.10Ϯ0.20 0.00Ϯ0.00

Data are presented as mean Ϯ SD (n ϭ 4). †, P Ͻ 0.05 compared with respective filtered air control. ‡, P Ͻ 0.05 compared with tobacco smoke plus vehicle. §, P Ͻ 0.05 compared with tobacco smoke plus AUDA-nBE. ¶, P Ͻ 0.05 compared with filtered air plus vehicle.

inflammation in the lung. This enzyme has Ͼ90% homology vascular , have a variety of antiinflammatory effects between rodent and human (26) and can be inhibited in vitro by including inhibiting expression of several molecules a number of urea, carbamate, and amide derivatives (27, 28). and inhibiting leukocyte adhesion to the vascular wall (12, 14, Injection of the sEH inhibitor N,NЈ-dicyclohexyl urea (DCU) in 29). The data in this study can be interpreted in light of these SH rats resulted in lower blood pressure, an increase in urinary hypothetical mechanisms of action. Oxylipid concentrations in 14(15)-EET, and a decrease in the corresponding urinary diol plasma may reflect local changes in the concentration of these (DHET). These observations are consistent with in vivo inhibi- important chemical mediators but are not equivalent to the local tion of sEH by DCU. AUDA and its n-butyl ester are more concentrations. The results certainly are complicated by changes potent sEH inhibitors than DCU. The blood levels of AUDA due to biosynthesis, distribution, and possibly other factors. To after s.c. administration of AUDA-nBE are high enough to fully appreciate the measure of AUDA-nBE and AUDA activity suggest that the sEH is significantly inhibited. with regard to attenuation of inflammation, some of the results In this study, we found pulmonary inflammation to be induced will be discussed in terms of epoxy:diol ratios. As expected, the in rats exposed to tobacco smoke for 3 days. Exposure to tobacco proinflammatory 12,13-DiHOME, 9,10-DiHOME, and 14,15- smoke significantly increased by 3.2-fold the total number of DHET increased dramatically by 5.1-, 2.0-, and 2.0-fold, respec- cells recovered by BAL. s.c. injection of AUDA-nBE signifi- tively, after exposure to tobacco smoke. After exposure to cantly decreased the total BAL cell number by 37% in tobacco tobacco smoke, epoxy:diol ratios in plasma were decreased by smoke-exposed rats. Numbers of BAL macrophages, neutro- 14.3-, 6.8-, and 2.0-fold for 12,13-EpHOME:DiHOME, 9,10- phils, and lymphocytes were also significantly increased with EpHOME:DiHOME, and 14,15-EET:DHET, respectively. tobacco smoke exposure, whereas AUDA-nBE significantly These effects are largely reversed by both treatments. With decreased numbers of these cell types by 18%, 55%, and 74%, administration of AUDA-nBE before exposure to tobacco respectively, in smoke-exposed animals. The combination of smoke, 12,13-EpHOME:DiHOME, 9,10-EpHOME:DiHOME, AUDA-nBE and EETs further reduced tobacco smoke-induced and 14,15-EET:DHET ratios increased by 15.9-, 1.9-, and 2.1- inflammation compared with AUDA-nBE alone. Interestingly, fold, respectively. With s.c. injection of AUDA-nBE before treatment of animals with both AUDA-nBE and EETs before smoke exposure, 12,13-EpHOME:DiHOME, 9,10-EpHO- exposure to filtered air resulted in significantly lower numbers of ME:DiHOME, and 14,15-EET:DHET ratios were 111%, 28%, total BAL cells and macrophages compared with animals treated and 104% respectively, of those in animals treated with vehicle only with vehicle before exposure to filtered air. This result may and exposed to filtered air. This effect is even greater with represent a basal effect of these compounds in ‘‘control’’ animals combined AUDA-nBE and EETs treatment. With coadminis- and may contribute to the reduced inflammation during expo- tration of AUDA-nBE and EETs before exposure to tobacco sure to tobacco smoke. smoke, 12,13-EpHOME:DiHOME, 9,10-EpHOME:DiHOME, It is hypothesized that AUDA acts by stabilizing antiinflam- and 14,15-EET:DHET ratios increased by 23.2-, 4.5-, and 2.7- matory fatty acid epoxides such as EETs and reducing the fold, respectively. With administration of both AUDA-nBE and formation of proinflammatory fatty acid diols such as the EETs before smoke exposure, 12,13-EpHOME:DiHOME, 9,10- DiHOMEs in the local environment where they are formed. At EpHOME:DiHOME, and 14,15-EET:DHET ratios were 162%, nanomolar concentrations, the EETs, which are produced by the 66%, and 135% respectively, of those in animals treated with

Table 2. Epoxy and diol oxylipid concentrations in rat plasma after 3 days of tobacco smoke exposure 12(13)- 9(10)- 12,13- 9,10- 14,15- 11,12- 8,9- 5,6- EpOME EpOME 14(15)-EET 11(12)-EET 8(9)-EET 5(6)-EET DiHOME DiHOME DHET DHET DHET DHET

Filtered air Vehicle 23.7Ϯ3.2 10.9Ϯ1.4 1.91Ϯ0.17 2.54Ϯ0.76 0.390Ϯ0.46 5.05Ϯ1.8 16.4Ϯ0.50 17.1Ϯ2.8 1.31Ϯ0.32 1.41Ϯ0.18 3.47Ϯ0.15 1.19Ϯ0.19 AUDA- 24.2Ϯ1.9 8.74Ϯ0.58† 2.28Ϯ0.46 0.670Ϯ1.2† 0.00Ϯ0.00 4.06Ϯ0.67 13.5Ϯ3.5 16.9Ϯ2.5 1.03Ϯ0.15 1.67Ϯ0.41 4.51Ϯ1.5 1.88Ϯ0.37† nBE AUDA- 29.2Ϯ7.5 9.49Ϯ2.2 2.76Ϯ0.35† 1.78Ϯ1.3 1.19Ϯ1.4 4.53Ϯ2.1 11.4Ϯ1.75† 15.7Ϯ1.5 1.41Ϯ0.32 2.87Ϯ1.1 3.73Ϯ0.38 1.64Ϯ0.17† nBE ϩ EETs Tobacco smoke Vehicle 8.49Ϯ1.2† 3.26Ϯ0.50† 1.85Ϯ0.20 1.60Ϯ0.31 2.70Ϯ0.84† 3.99Ϯ0.20 84.1Ϯ33† 34.7Ϯ11† 2.56Ϯ0.74† 1.91Ϯ0.22† 4.95Ϯ0.57† 2.01Ϯ0.23† AUDA- 32.0Ϯ7.1‡ 5.13Ϯ1.1†‡ 2.39Ϯ0.28†‡ 1.85Ϯ0.27 2.69Ϯ0.76† 4.13Ϯ0.79 19.9Ϯ0.89†‡ 28.7Ϯ1.9† 1.58Ϯ0.32 2.00Ϯ0.22† 5.17Ϯ0.65† 2.09Ϯ0.13† nBE AUDA- 35.1Ϯ2.2†‡ 8.61Ϯ1.4‡ 2.65Ϯ0.67 2.60Ϯ0.37‡ 2.39Ϯ1.94 4.53Ϯ1.2 15.0Ϯ0.65†‡ 20.5Ϯ1.9‡ 1.35Ϯ0.11‡ 3.02Ϯ0.49†‡ 4.58Ϯ0.85 1.23Ϯ0.17‡ nBE ϩ EETs

Data are presented in nM as mean Ϯ SD (n ϭ 3–4). †, P Ͻ 0.05 compared with filtered air and vehicle. ‡, P Ͻ 0.05 compared with tobacco smoke and vehicle (only within tobacco smoke groups).

2190 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0409591102 Smith et al. Downloaded by guest on October 2, 2021 vehicle and exposed to filtered air. Because 12(13)- and 9(10)- induced inflammation are attributable solely to an increase in EpOME were not in the wax plug, the mechanism for their EETs levels in our animal model is unknown, although the overall enhancement by EET administration is not clear. The further reduction in inflammation by coadministration of EETs 11,12-EET:DHET ratio did not follow the trend of the linoleate and AUDA-nBE supports this hypothesis. epoxy:diol ratios because of a significant increase in diol con- In summary, s.c. administration of AUDA-nBE, in the pres- centrations with coadministration of AUDA-nBE and EETs. ence or absence of EETs, caused marked reduction in tobacco 8(9)-EET was increased in animals after exposure to tobacco smoke-induced inflammation when given before and during smoke and did not change significantly with AUDA-nBE treat- smoke exposure. Our studies suggest that pharmacologic inhi- ment alone or in combination with EETs, possibly due to bition of sEH with or without the coadministration of EETs shuttling of 8(9)-EET through the pathway when might be a mechanism to modulate pulmonary epoxy and diol sEH was inhibited. 5(6)-EET is the least preferred substrate for levels for altering inflammation commonly associated with such sEH and was not included in the EETs treatment because of common human disorders as COPD and carcinogenesis. lactone formation, so it was not surprising that there were no significant differences between the six treatment groups. How- We thank Kara Schmelzer for advice on oxylipid extraction. This work ever, the 5,6-DHET concentration was significantly increased by was supported in part by Grants R01 ES011634 and R37 ES02710, the exposure to tobacco smoke, and this smoke-induced increase Superfund Basic Research Program (P42 ES04699), the Environmental was attenuated in animals treated with the combination of Health Sciences Center (P30 ES05707), and the Center for Children’s AUDA-nBE and EETs. Because of the high polarity of these Environmental Health and Disease Prevention (1 P01 ES11269), all from diols and the ease of conjugate formation, systemic plasma the National Institute of Environmental Health Sciences; and the concentrations probably underestimate the importance of this University of California Systemwide Biotechnology Research and Ed- pathway. Whether the effects of AUDA-nBE on tobacco smoke- ucation Training Grant 2001-07.

1. Snider, G. L. (1996) in Pulmonary and Critical Care Pharmacology and 16. Sinal, C. J., Miyata, M., Tohkin, M., Nagata, K., Bend, J. R. & Gonzalez, F. J. Therapeutics., ed. Leff, A. R. (McGraw-Hill, New York), pp. 821–828. (2000) J. Biol. Chem. 275, 40504–40510. 2. Jeffery, P. K. (1998) Thorax 53, 129–136. 17. Smith, K. R., Uyeminami, D. L., Kodavanti, U. P., Crapo, J. D., Chang, L. Y. 3. Fournier, M., Lebargy, F., Le Roy Ladurie, F., Lenormand, E. & Pariente, R. & Pinkerton, K. E. (2002) Free Radical Biol. Med. 33, 1106–1114. (1989) Am. Rev. Respir. Dis. 140, 737–742. 18. Newman, J. W., Stok, J. E., Vidal, J. D., Corbin, C. J., Huang, Q., Hammock, 4. Saetta, M., Turato, G., Facchini, F. M., Corbino, L., Lucchini, R. E., Casoni, B. D. & Conley, A. J. (2004) Endocrinology 145, 5097–5105. G., Maestrelli, P., Mapp, C. E., Ciaccia, A. & Fabbri, L. M. (1997) Am. J. Respir. 19. Newman, J. W., Watanabe, T. & Hammock, B. D. (2002) J. Lipid Res. 43, Crit. Care Med. 156, 1633–1639. 1563–1578. 5. Grashoff, W. F., Sont, J. K., Sterk, P. J., Hiemstra, P. S., de Boer, W. I., Stolk, 20. Zhang, Y. E. & Schwartz, J. B. (2003) Drug Dev. Ind. Pharm. 29, 131–138. J., Han, J. & van Krieken, J. M. (1997) Am. J. Pathol. 151, 1785–1790. 21. Teague, S. V., Pinkerton, K. E., Goldsmith, M., Gebremichael, A., Chang, S., 6. Saetta, M., Di Stefano, A., Turato, G., Facchini, F. M., Corbino, L., Mapp, Jenkins, R. A. & Moneyhun, J. H. (1994) Inhal. Toxicol. 6, 79–93. C. E., Maestrelli, P., Ciaccia, A. & Fabbri, L. M. (1998) Am. J. Respir. Crit. Care 22. Gossart, S., Cambon, C., Orfila, C., Seguelas, M. H., Lepert, J. C., Rami, J., Med. 157, 822–826. Carre, P. & Pipy, B. (1996) J. Immunol. 156, 1540–1548. 7. Hammock, B. D., Storms, D. H. & Grant, D. F. (1997) in Comprehensive 23. Wood-Baker, R. (2003) Am. J. Respir. Med. 2, 451–458. Toxicology: Biotransformation (Elsevier, New York), pp. 283–305.

24. Zeldin, D. C., Kobayashi, J., Falck, J. R., Winder, B. S., Hammock, B. D., PHARMACOLOGY 8. Oesch, F. (1972) Xenobiotica 3, 305–340. Snapper, J. R. & Capdevila, J. H. (1993) J. Biol. Chem. 268, 6402–6407. 9. Moghaddam, M. F., Grant, D. F., Cheek, J. M., Greene, J. F., Williamson, K. C. 25. Fang, X., Kaduce, T. L., Weintraub, N. L., Harmon, S., Teesch, L. M., & Hammock, B. D. (1997) Nat. Med. 3, 562–566. Morisseau, C., Thompson, D. A., Hammock, B. D. & Spector, A. A. (2001) 10. Slim, R., Hammock, B. D., Toborek, M., Robertson, L. W., Newman, J. W., Morisseau, C. H., Watkins, B. A., Saraswathi, V. & Hennig, B. (2001) Toxicol. J. Biol. Chem. 276, 14867–14874. Appl. Pharmacol. 171, 184–193. 26. Arand, M., Grant, D. F., Beetham, J. K., Friedberg, T., Oesch, F. & Hammock, 11. Zheng, J., Plopper, C. G., Lakritz, J., Storms, D. H. & Hammock, B. D. (2001) B. D. (1994) FEBS Lett. 338, 251–256. Am. J. Respir. Cell Mol. Biol. 25, 434–438. 27. Morisseau, C., Goodrow, M. H., Dowdy, D., Zheng, J., Greene, J. F., 12. Campbell, W. B. (2000) Trends Pharmacol. Sci. 21, 125–127. Sanborn, J. R. & Hammock, B. D. (1999) Proc. Natl. Acad. Sci. USA 96, 13. Zeldin, D. C. & Liao, J. K. (2000) Trends Pharmacol. Sci. 21, 127–128. 8849–8854. 14. Node, K., Huo, Y., Ruan, X., Yang, B., Spiecker, M., Ley, K., Zeldin, D. C. & 28. Morisseau, C., Goodrow, M. H., Newman, J. W., Wheelock, C. E., Dowdy, D. L. Liao, J. K. (1999) Science 285, 1276–1279. & Hammock, B. D. (2002) Biochem. Pharmacol. 63, 1599–1608. 15. Yu, Z., Xu, F., Huse, L. M., Morisseau, C., Draper, A. J., Newman, J. W., 29. Falck, J. R., Reddy, L. M., Reddy, Y. K., Bondlela, M., Krishna, U. M., Parker, C., Graham, L., Engler, M. M., Hammock, B. D., et al. (2000) Circ. Res. Ji, Y., Sun, J. & Liao, J. K. (2003) Bioorg. Med. Chem. Lett. 13, 4011– 87, 992–998. 4014.

Smith et al. PNAS ͉ February 8, 2005 ͉ vol. 102 ͉ no. 6 ͉ 2191 Downloaded by guest on October 2, 2021